Large data sets may exist in various levels of size and organization. With big data comprising data sets as large as ever, the volume of data collected incident to the increased popularity of online and electronic transactions continues to grow. Billions of rows and hundreds of thousands of columns worth of data may populate a single table, for example. Different users may desire to access different portions of the large volume of data populating the table. However, different portions of the table may have different security restrictions, for instance, some portions may contain personally identifiable information that should have higher security. Moreover, various portions of the table and/or the entire table may be encrypted at rest. Different users may have different permissions specific to their roles, identities, and the row and/or column of the table desired to be accessed. However, the size and varied structure of big data sets is typically incompatible with traditional data protection techniques.